All of lore.kernel.org
 help / color / mirror / Atom feed
From: David Gibson <david@gibson.dropbear.id.au>
To: matheus.ferst@eldorado.org.br
Cc: richard.henderson@linaro.org, qemu-devel@nongnu.org,
	f4bug@amsat.org, luis.pires@eldorado.org.br, qemu-ppc@nongnu.org,
	lagarcia@br.ibm.com, bruno.larsen@eldorado.org.br
Subject: Re: [PATCH v5 20/23] target/ppc: Implement cfuged instruction
Date: Tue, 18 May 2021 10:51:07 +1000	[thread overview]
Message-ID: <YKMPe0n3H5fh/3zl@yekko> (raw)
In-Reply-To: <20210517205025.3777947-21-matheus.ferst@eldorado.org.br>

[-- Attachment #1: Type: text/plain, Size: 5334 bytes --]

On Mon, May 17, 2021 at 05:50:22PM -0300, matheus.ferst@eldorado.org.br wrote:
> From: Matheus Ferst <matheus.ferst@eldorado.org.br>
> 
> Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>

Applied to ppc-for-6.1, thanks.

> ---
> v5:
> - Comments to explain helper_cfuged implementation.
> ---
>  target/ppc/helper.h                        |  1 +
>  target/ppc/insn32.decode                   |  4 ++
>  target/ppc/int_helper.c                    | 62 ++++++++++++++++++++++
>  target/ppc/translate/fixedpoint-impl.c.inc | 12 +++++
>  4 files changed, 79 insertions(+)
> 
> diff --git a/target/ppc/helper.h b/target/ppc/helper.h
> index ea9f2a236c..c517b9f025 100644
> --- a/target/ppc/helper.h
> +++ b/target/ppc/helper.h
> @@ -46,6 +46,7 @@ DEF_HELPER_4(divwe, tl, env, tl, tl, i32)
>  DEF_HELPER_FLAGS_1(popcntb, TCG_CALL_NO_RWG_SE, tl, tl)
>  DEF_HELPER_FLAGS_2(cmpb, TCG_CALL_NO_RWG_SE, tl, tl, tl)
>  DEF_HELPER_3(sraw, tl, env, tl, tl)
> +DEF_HELPER_FLAGS_2(cfuged, TCG_CALL_NO_RWG_SE, i64, i64, i64)
>  #if defined(TARGET_PPC64)
>  DEF_HELPER_FLAGS_2(cmpeqb, TCG_CALL_NO_RWG_SE, i32, tl, tl)
>  DEF_HELPER_FLAGS_1(popcntw, TCG_CALL_NO_RWG_SE, tl, tl)
> diff --git a/target/ppc/insn32.decode b/target/ppc/insn32.decode
> index bc69c70493..d4044d9069 100644
> --- a/target/ppc/insn32.decode
> +++ b/target/ppc/insn32.decode
> @@ -87,6 +87,10 @@ STDUX           011111 ..... ..... ..... 0010110101 -   @X
>  ADDI            001110 ..... ..... ................     @D
>  ADDIS           001111 ..... ..... ................     @D
>  
> +## Fixed-Point Logical Instructions
> +
> +CFUGED          011111 ..... ..... ..... 0011011100 -   @X
> +
>  ### Move To/From System Register Instructions
>  
>  SETBC           011111 ..... ..... ----- 0110000000 -   @X_bi
> diff --git a/target/ppc/int_helper.c b/target/ppc/int_helper.c
> index 41f8477d4b..efa833ef64 100644
> --- a/target/ppc/int_helper.c
> +++ b/target/ppc/int_helper.c
> @@ -320,6 +320,68 @@ target_ulong helper_popcntb(target_ulong val)
>  }
>  #endif
>  
> +uint64_t helper_cfuged(uint64_t src, uint64_t mask)
> +{
> +    /*
> +     * Instead of processing the mask bit-by-bit from the most significant to
> +     * the least significant bit, as described in PowerISA, we'll handle it in
> +     * blocks of 'n' zeros/ones from LSB to MSB. To avoid the decision to use
> +     * ctz or cto, we negate the mask at the end of the loop.
> +     */
> +    target_ulong m, left = 0, right = 0;
> +    unsigned int n, i = 64;
> +    bool bit = false; /* tracks if we are processing zeros or ones */
> +
> +    if (mask == 0 || mask == -1) {
> +        return src;
> +    }
> +
> +    /* Processes the mask in blocks, from LSB to MSB */
> +    while (i) {
> +        /* Find how many bits we should take */
> +        n = ctz64(mask);
> +        if (n > i) {
> +            n = i;
> +        }
> +
> +        /*
> +         * Extracts 'n' trailing bits of src and put them on the leading 'n'
> +         * bits of 'right' or 'left', pushing down the previously extracted
> +         * values.
> +         */
> +        m = (1ll << n) - 1;
> +        if (bit) {
> +            right = ror64(right | (src & m), n);
> +        } else {
> +            left = ror64(left | (src & m), n);
> +        }
> +
> +        /*
> +         * Discards the processed bits from 'src' and 'mask'. Note that we are
> +         * removing 'n' trailing zeros from 'mask', but the logical shift will
> +         * add 'n' leading zeros back, so the population count of 'mask' is kept
> +         * the same.
> +         */
> +        src >>= n;
> +        mask >>= n;
> +        i -= n;
> +        bit = !bit;
> +        mask = ~mask;
> +    }
> +
> +    /*
> +     * At the end, right was ror'ed ctpop(mask) times. To put it back in place,
> +     * we'll shift it more 64-ctpop(mask) times.
> +     */
> +    if (bit) {
> +        n = ctpop64(mask);
> +    } else {
> +        n = 64 - ctpop64(mask);
> +    }
> +
> +    return left | (right >> n);
> +}
> +
>  /*****************************************************************************/
>  /* PowerPC 601 specific instructions (POWER bridge) */
>  target_ulong helper_div(CPUPPCState *env, target_ulong arg1, target_ulong arg2)
> diff --git a/target/ppc/translate/fixedpoint-impl.c.inc b/target/ppc/translate/fixedpoint-impl.c.inc
> index 204848d017..4038143efb 100644
> --- a/target/ppc/translate/fixedpoint-impl.c.inc
> +++ b/target/ppc/translate/fixedpoint-impl.c.inc
> @@ -227,3 +227,15 @@ TRANS(SETBC, do_set_bool_cond, false, false)
>  TRANS(SETBCR, do_set_bool_cond, false, true)
>  TRANS(SETNBC, do_set_bool_cond, true, false)
>  TRANS(SETNBCR, do_set_bool_cond, true, true)
> +
> +static bool trans_CFUGED(DisasContext *ctx, arg_X *a)
> +{
> +    REQUIRE_64BIT(ctx);
> +    REQUIRE_INSNS_FLAGS2(ctx, ISA310);
> +#if defined(TARGET_PPC64)
> +    gen_helper_cfuged(cpu_gpr[a->ra], cpu_gpr[a->rt], cpu_gpr[a->rb]);
> +#else
> +    qemu_build_not_reached();
> +#endif
> +    return true;
> +}

-- 
David Gibson			| I'll have my music baroque, and my code
david AT gibson.dropbear.id.au	| minimalist, thank you.  NOT _the_ _other_
				| _way_ _around_!
http://www.ozlabs.org/~dgibson

[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

  reply	other threads:[~2021-05-18  1:53 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-17 20:50 [PATCH v5 00/23] Base for adding PowerPC 64-bit instructions matheus.ferst
2021-05-17 20:50 ` [PATCH v5 01/23] target/ppc: Introduce gen_icount_io_start matheus.ferst
2021-05-18  0:13   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 02/23] target/ppc: Replace POWERPC_EXCP_STOP with DISAS_EXIT_UPDATE matheus.ferst
2021-05-18  0:14   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 03/23] target/ppc: Replace POWERPC_EXCP_BRANCH with DISAS_NORETURN matheus.ferst
2021-05-18  0:15   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 04/23] target/ppc: Remove DisasContext.exception matheus.ferst
2021-05-18  0:17   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 05/23] target/ppc: Move single-step check to ppc_tr_tb_stop matheus.ferst
2021-05-18  0:19   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 06/23] target/ppc: Tidy exception vs exit_tb matheus.ferst
2021-05-18  0:19   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 07/23] target/ppc: Mark helper_raise_exception* as noreturn matheus.ferst
2021-05-18  0:20   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 08/23] target/ppc: Use translator_loop_temp_check matheus.ferst
2021-05-18  0:20   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 09/23] target/ppc: Introduce macros to check isa extensions matheus.ferst
2021-05-18  0:21   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 10/23] target/ppc: Move page crossing check to ppc_tr_translate_insn matheus.ferst
2021-05-18  0:23   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 11/23] target/ppc: Add infrastructure for prefixed insns matheus.ferst
2021-05-18  0:25   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 12/23] target/ppc: Move ADDI, ADDIS to decodetree, implement PADDI matheus.ferst
2021-05-18  0:35   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 13/23] target/ppc: Implement PNOP matheus.ferst
2021-05-18  0:36   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 14/23] TCG: add tcg_constant_tl matheus.ferst
2021-05-18  0:37   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 15/23] target/ppc: Move D/DS/X-form integer loads to decodetree matheus.ferst
2021-05-18  0:44   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 16/23] target/ppc: Implement prefixed integer load instructions matheus.ferst
2021-05-18  0:45   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 17/23] target/ppc: Move D/DS/X-form integer stores to decodetree matheus.ferst
2021-05-18  0:47   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 18/23] target/ppc: Implement prefixed integer store instructions matheus.ferst
2021-05-18  0:47   ` David Gibson
2021-05-17 20:50 ` [PATCH v5 19/23] target/ppc: Implement setbc/setbcr/stnbc/setnbcr instructions matheus.ferst
2021-05-18  0:49   ` David Gibson
2021-05-18  9:48   ` Richard Henderson
2021-05-17 20:50 ` [PATCH v5 20/23] target/ppc: Implement cfuged instruction matheus.ferst
2021-05-18  0:51   ` David Gibson [this message]
2021-05-17 20:50 ` [PATCH v5 21/23] target/ppc: Implement vcfuged instruction matheus.ferst
2021-05-18  0:52   ` David Gibson
2021-05-18  9:54   ` Richard Henderson
2021-05-17 20:50 ` [PATCH v5 22/23] target/ppc: Move addpcis to decodetree matheus.ferst
2021-05-18  0:53   ` David Gibson
2021-05-18  9:55   ` Richard Henderson
2021-05-17 20:50 ` [PATCH v5 23/23] target/ppc: Move cmp/cmpi/cmpl/cmpli " matheus.ferst
2021-05-18  0:56   ` David Gibson
2021-05-18 10:12   ` Richard Henderson
2021-05-21 17:25     ` Matheus K. Ferst
2021-05-24 18:51       ` Richard Henderson
2021-05-26 15:17         ` Matheus K. Ferst
2021-05-26 16:11           ` Richard Henderson
2021-05-27  1:11           ` David Gibson
2021-05-18  3:58 ` [PATCH v5 00/23] Base for adding PowerPC 64-bit instructions David Gibson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=YKMPe0n3H5fh/3zl@yekko \
    --to=david@gibson.dropbear.id.au \
    --cc=bruno.larsen@eldorado.org.br \
    --cc=f4bug@amsat.org \
    --cc=lagarcia@br.ibm.com \
    --cc=luis.pires@eldorado.org.br \
    --cc=matheus.ferst@eldorado.org.br \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@nongnu.org \
    --cc=richard.henderson@linaro.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.